Filters, Contrast, Transformation and Morphology

• Jump into filtering
  • Edge detection
  • Blurring to reduce noise
• Contrast enhancement
  • Medical images
  • Aerial image
  • Let's add some impact and contrast
• Transformations
  • Aliasing, rotating and rescaling
  • Enlarging images
  • Proportionally resizing
• Morphology
  • Handwritten letters
  • Improving thresholded image

import numpy as np
import matplotlib.pyplot as plt

Jump into filtering

• Filters
  • Enhancing an image
  • Emphasize or remove features
  • Smoothing
  • Sharpening
  • Edge detection

Edge detection

In this exercise, you'll detect edges in an image by applying the Sobel filter.

def show_image(image, title='Image', cmap_type='gray'):
    plt.imshow(image, cmap=cmap_type)
    plt.title(title)
    plt.axis('off')

soaps_image = plt.imread('./dataset/soap_image.jpg')

from skimage import color
from skimage.filters import sobel

# Make the image grayscale
soaps_image_gray = color.rgb2gray(soaps_image)

# apply edge detection filters
edge_sobel = sobel(soaps_image_gray)

# Show original image
show_image(soaps_image, 'Original')

show_image(edge_sobel, 'Edges with Sobel')

Blurring to reduce noise

In this exercise you will reduce the sharpness of an image of a building taken during a London trip, through filtering.

from skimage.filters import gaussian

building_image = plt.imread('./dataset/building_image.jpg')

# Apply filter
gaussian_image = gaussian(building_image, multichannel=True)

# Show the original image
show_image(building_image, 'Original')

show_image(gaussian_image, 'Reduced sharpness Gaussian')

Contrast enhancement

• Contrast
  • difference between the maximum and minimum pixel intensity in the image
• Enhance contrast
  • Contrast stretching
  • Histogram equalization

Medical images

You are trying to improve the tools of a hospital by pre-processing the X-ray images so that doctors have a higher chance of spotting relevant details. You'll test our code on a chest X-ray image from the National Institutes of Health Chest X-Ray Dataset

First, you'll check the histogram of the image and then apply standard histogram equalization to improve the contrast.

from skimage import exposure

chest_xray_image = plt.imread('./dataset/chest_xray_image.png')

# Show original x-ray image and its histogram
show_image(chest_xray_image, 'Original x-ray')

plt.title('Histogram of image')
plt.hist(chest_xray_image.ravel(), bins=256);

xray_image_eq = exposure.equalize_hist(chest_xray_image)

# Show the resulting image
show_image(xray_image_eq, 'Resulting image')

Aerial image

In this exercise, we will improve the quality of an aerial image of a city. The image has low contrast and therefore we can not distinguish all the elements in it.

For this we will use the normal or standard technique of Histogram Equalization.

image_aerial = plt.imread('./dataset/image_aerial.png')

# Use histogram equalization to improve the contrast
image_eq = exposure.equalize_hist(image_aerial)

# Show the original image
show_image(image_aerial, 'Original')

show_image(image_eq, 'Resulting image')

Let's add some impact and contrast

Have you ever wanted to enhance the contrast of your photos so that they appear more dramatic?

In this exercise, you'll increase the contrast of a cup of coffee.

Even though this is not our Sunday morning coffee cup, you can still apply the same methods to any of our photos.

from skimage import data

# Load the image 
original_image = data.coffee()

# Apply the adaptive equalization on the original image
adapthist_eq_image = exposure.equalize_adapthist(original_image, clip_limit=0.03)

# Compare the original image
show_image(original_image)